Ollama fails to load this

#4
by raymondz1 - opened

When Ollama loads the GGUF, it expects a specific list of tensors; because it does not recognize or find  output_norm , it aborts with  error loading model: missing tensor 'output_norm'  and returns HTTP 500

Liquid AI org

Please submit the issue to ollama, GGUFs work with the latest b7622 llama.cpp release

image

tarek-liquid changed discussion status to closed

Sign up or log in to comment